MIPS: UAPI: Fix unrecognized opcode WSBH/DSBH/DSHD when using MIPS16.

The nomips16 has to be added both as function attribute and assembler
directive.

When only function attribute is specified, the compiler will inline the
function with -Os optimization.  The generated assembly code cannot be
correctly assembled because ISA mode switch has to be done through jump
instruction.

When only ".set nomips16" directive is used, the generated assembly code
will use MIPS32 code for the inline assembly template and MIPS16 for the
function return.  The compiled binary is invalid:

    00403100 <__arch_swab16>:
      403100:   7c0410a0    wsbh    v0,a0
      403104:   e820ea31    swc2    $0,-5583(at)

while correct code should be:

    00402650 <__arch_swab16>:
      402650:   7c0410a0    wsbh    v0,a0
      402654:   03e00008    jr  ra
      402658:   3042ffff    andi    v0,v0,0xffff

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
Cc: Chen Jie <chenj@lemote.com>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/11087/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

diff --git a/arch/mips/include/uapi/asm/swab.h b/arch/mips/include/uapi/asm/swab.h
index 8f2d184dbe9f7adb43a32325e0061243c8f5fbf9..c4ddc4f0d2dcb11c7aa55167434d72e7990e083a 100644 (file)
--- a/arch/mips/include/uapi/asm/swab.h
+++ b/arch/mips/include/uapi/asm/swab.h
@@ -16,11 +16,13 @@
 #if (defined(__mips_isa_rev) && (__mips_isa_rev >= 2)) ||              \
     defined(_MIPS_ARCH_LOONGSON3A)
 
 #if (defined(__mips_isa_rev) && (__mips_isa_rev >= 2)) ||              \
     defined(_MIPS_ARCH_LOONGSON3A)
 

-static inline __attribute_const__ __u16 __arch_swab16(__u16 x)
+static inline __attribute__((nomips16)) __attribute_const__
+               __u16 __arch_swab16(__u16 x)

 {
        __asm__(
        "       .set    push                    \n"
        "       .set    arch=mips32r2           \n"
 {
        __asm__(
        "       .set    push                    \n"
        "       .set    arch=mips32r2           \n"

+       "       .set    nomips16                \n"

        "       wsbh    %0, %1                  \n"
        "       .set    pop                     \n"
        : "=r" (x)
        "       wsbh    %0, %1                  \n"
        "       .set    pop                     \n"
        : "=r" (x)


@@ -30,11 +32,13 @@ static inline __attribute_const__ __u16 __arch_swab16(__u16 x)
 }
 #define __arch_swab16 __arch_swab16
 
 }
 #define __arch_swab16 __arch_swab16
 

-static inline __attribute_const__ __u32 __arch_swab32(__u32 x)
+static inline __attribute__((nomips16)) __attribute_const__
+               __u32 __arch_swab32(__u32 x)

 {
        __asm__(
        "       .set    push                    \n"
        "       .set    arch=mips32r2           \n"
 {
        __asm__(
        "       .set    push                    \n"
        "       .set    arch=mips32r2           \n"

+       "       .set    nomips16                \n"

        "       wsbh    %0, %1                  \n"
        "       rotr    %0, %0, 16              \n"
        "       .set    pop                     \n"
        "       wsbh    %0, %1                  \n"
        "       rotr    %0, %0, 16              \n"
        "       .set    pop                     \n"


@@ -50,11 +54,13 @@ static inline __attribute_const__ __u32 __arch_swab32(__u32 x)
  * 64-bit kernel on r2 CPUs.
  */
 #ifdef __mips64
  * 64-bit kernel on r2 CPUs.
  */
 #ifdef __mips64

-static inline __attribute_const__ __u64 __arch_swab64(__u64 x)
+static inline __attribute__((nomips16)) __attribute_const__
+               __u64 __arch_swab64(__u64 x)

 {
        __asm__(
        "       .set    push                    \n"
        "       .set    arch=mips64r2           \n"
 {
        __asm__(
        "       .set    push                    \n"
        "       .set    arch=mips64r2           \n"

+       "       .set    nomips16                \n"

        "       dsbh    %0, %1                  \n"
        "       dshd    %0, %0                  \n"
        "       .set    pop                     \n"
        "       dsbh    %0, %1                  \n"
        "       dshd    %0, %0                  \n"
        "       .set    pop                     \n"